Performance of universal machine learning potentials in global optimization

TL;DR

This work examined the latest generation of uMLPs in unconstrained evolutionary searches to assess whether these models can consistently predict complex crystal structure ground states across diverse inorganic systems.

Abstract

Rapid development of universal machine learning potentials (uMLPs) and expansion of training data sets are reshaping the state of the art in atomistic simulation, highlighting the need for concurrent systematic benchmarking of their capabilities. Global optimization is among the most demanding uMLP applications because unconstrained exploration includes probing motifs not present in reference sets. We examined the latest generation of uMLPs in unconstrained evolutionary searches to assess whether these models can consistently predict complex crystal structure ground states across diverse inorganic systems. Our findings demonstrate that the considered M3GNet, MACE, SevenNet, EquiformerV2, MatterSim, GRACE, eSEN, Orb-v3, and PET-MAD models span a wide performance range, from near ab initio to essentially non-predictive, in their ability to resolve competing phases within low-energy basins. Additional tests on hcp-Zn, MB$_4$ (M = Cr, Mn, and Fe), and LiB$_{y}$ ($y\approx 0.9$) ground states reveal that several uMLPs capture fine energy differences arising from subtle electronic structure features.

Figures (6)

• Figure 1: Stability of the tI10-Na$_2$CN$_2$, mS28-MgB$_3$C$_3$, and oI28-MgB$_3$C$_3$ phases identified in this work relative to reported mS10-Na$_2$CN$_2$ and hP14-MgB$_3$C$_3$ evaluated with common DFT functionals.
• Figure 2: Performance metrics on merged pools assessed relative to the reference DFT method, PBEsol for PT and PBE for the rest, and averaged over 11 compounds, excluding AgClO$_4$. The schematics at the top clarify the definitions of the proximity and ranking metrics introduced in the text. The representative merged EN pools were used to evaluate the metrics for the alternative DFT approximations and MG, with dashed horizontal lines marking the ranking RMSE for PBEsol and r$^2$SCAN relative to PBE. The energy proximity average for MG exceeded 70 meV/atom (see Fig. S2 for further discussion).
• Figure 3: Band structure and density of states calculated with DFT-PBE for hcp-Zn (black) and fcc-Zn (red) represented with 6-atom hexagonal unit cells at the experimental $c/a = 1.826$ ratio.
• Figure 4: DFT and MLP energy profiles of hcp-Zn as a function of $c/a$ for fixed-volume unit cells. Each curve is set to zero at the experimental $c/a = 1.826$ ratio.
• Figure 5: Stability of distorted oP10 and mP20 derivatives relative to oI10 calculated with DFT and uMLPs for MB$_4$ with (a) M = Cr, (b) M = Mn, and (c) M = Fe. The oI10 and oP10 structures are displayed with supercells to illustrate the connection with the lowest-symmetry mP20.